Adult rhesus monkeys (Macaca mulata) were vaccinated with four inactivated rabies vaccines, including two cell culture vaccines, one zonal purified cell culture vaccine, and a 10% extracted duck embryo vaccine. The vaccines were potency tested by both National Institutes of Health (NIH) and Habel methods and passed one or both tests. However, a vaccine having acceptable potency by one method frequently failed or was marginal by the other procedure. Groups of three monkeys were inoculated with each vaccine by one of two schedules. The first consisted of four weekly 1-ml doses followed by a 1-ml booster dose at 6 months, and the second consisted of seven daily 1-ml doses of vaccine with no booster. Both zonal purified and extracted duck embryo vaccines induced detectable neutralizing antibody by day 7 with either schedule, and antibody titers elicited by the cell culture vaccine remained high through 210 days. However, antibody titers produced by the 10% duck embryo vaccine dropped sharply after their 28-day peak. Duck embryo cell culture vaccines with low or marginal potency as measured by Habel or NIH tests still produced rapid, high levels of serum-neutralizing antibody in primates. LD(50) or NIH and Habel tests as measured in mice were not necessarily good indices of antibody response in the primate host. The need for a cell culture potency test that will yield a more predictable correlation with the definitive host's antibody response is discussed.